May 17, 2006 The Honorable T.

Kenneth James Commissioner Arkansas Department of Education Four State Capitol Mall, Room 304 A Little Rock, AR 72201-1071 Dear Commissioner James: Thank you for submitting a proposal for the U.S. Department of Education’s (Department) growth-based accountability model pilot project. I realize that our timelines were tight and sometimes inconvenient; I appreciate the work you and your staff have done to participate in this effort so far. The Department continues to believe that this pilot project can help determine whether growth models will, most importantly, provide a fair, reliable, and innovative mechanism for holding schools accountable for ensuring that all students reach grade-level standards in reading and mathematics by 2013–14. In mid-March, the Department determined that Arkansas was meeting the “bright line principles” of the law – that is, ensuring all students are learning, making the system accountable, providing information and offering parents options, and improving teacher quality – and that the Arkansas growth model proposal seemed poised to meet the seven core principles outlined by Secretary Spellings in her letter on November 21, 2005. As such, Arkansas’ proposal was forwarded to a group of peer reviewers who met on April 17–18, 2006. The range of changes and the number of conditions that the peer reviewers indicated would be required for Arkansas’ model to be acceptable would be tantamount to writing a new proposal. On that basis, the Department is not approving the Arkansas proposal. However, with this letter I am inviting Arkansas to consider the peer reviewers’ feedback and submit a revised proposal by September 15, 2006. Our intent is that the Department will again conduct an initial review and advance acceptable proposals to a second peer review to take place in mid-October that will be organized solely for Arkansas and the five other States who have advanced to this point in the process. If successful, Arkansas’ growth model could be approved for implementation for the 2006– 07 school year. To help you with this effort, in addition to the information in this letter, I am providing two pieces of information: 1) the peer report for Arkansas; and 2) a document produced by the peer review team that outlines several general themes and cross-cutting concerns raised during the peer review, although not necessarily specific to

Page 2 – The Honorable T. Kenneth James Arkansas’ proposal. My staff and I are willing to discuss this information with you to help refine Arkansas’ proposal. The peers identified several strengths in the Arkansas proposal: the growth model included result from students who were taking the alternate assessment based on alternate achievement standards and the overall proposal demonstrated a commitment to use accountability to improve student achievement. However, the peers raised significant specific concerns regarding the Arkansas proposal. (Please refer to the enclosed peer report for details.) The peers were particularly concerned about Arkansas’ capacity, due to the implementation of a new tracking system, to match students’ records from one year to the next (a necessary step in calculating growth), as there were inconsistencies in how the data system was described and given the high percentage of unmatched students. A related concern was with the State’s untested plans to impute missing data for students. In addition, because the growth targets are reset annually, the peers felt that some students would never be expected to reach proficiency. In effect, some students would be counted as “proficient” (i.e., meeting the growth target) every year but never actually achieve grade-level standards. As a result, the growth model proposal did not support the goal of ensuring all students reach grade-level standards in reading and mathematics by 2013–14. The peers also expressed concern that the model that did not account for both proficient and non-proficient students, possibly ignoring the growth trajectories of proficient students. Empirical data from several States suggest this is a problem because a substantial number of students who are currently achieving on grade level do not seem poised to continue achieving at such a level in future grades. The peers were willing to entertain a model that calculated growth for all students even if the trajectories were not used for all students in adequate yearly progress (AYP) determinations; however, their other concerns with Arkansas’ proposal outweighed their interest in this approach. These and additional concerns are presented in the peer report that is enclosed. If Arkansas decides not to revise its current proposal based on this feedback and submit revised proposal by September 15, please note that the Department would welcome a new proposal from Arkansas along with other States later this year; these proposals would be due to us by November 1, 2006. The limit of approved plans through this pilot, however, will remain at ten. The Department will rigorously evaluate the approved proposals, review information on how each pilot project is working, and share the results with other States, policymakers, and the public. With the knowledge gained from the approved growth models, the Department will be able to make an informed decision on whether to expand the pilot project beyond the 2006–07 school year.

Page 3 – The Honorable T. Kenneth James Again, I appreciate your interest in the growth model pilot project and your continued efforts to ensure quality education for all children. Sincerely,

Henry L. Johnson Enclosures cc: Governor Mike Huckabee